Endocuff With or Without Artificial Intelligence-Assisted Colonoscopy in Detection of Colorectal Adenoma: A Randomized Colonoscopy Trial.

INTRODUCTION: Both artificial intelligence (AI) and distal attachment devices have been shown to improve adenoma detection rate and reduce miss rate during colonoscopy. We studied the combined effect of Endocuff and AI on enhancing detection rates of various colonic lesions. METHODS: This was a 3-arm prospective randomized colonoscopy study involving patients aged 40 years or older. Participants were randomly assigned in a 1:1:1 ratio to undergo Endocuff with AI, AI alone, or standard high-definition (HD) colonoscopy. The primary outcome was adenoma detection rate (ADR) between the Endocuff-AI and AI groups while secondary outcomes included detection rates of polyp (PDR), sessile serrated lesion (sessile detection rate [SDR]), and advanced adenoma (advanced adenoma detection rate) between the 2 groups. RESULTS: A total of 682 patients were included (mean age 65.4 years, 52.3% male), with 53.7% undergoing diagnostic colonoscopy. The ADR for the Endocuff-AI, AI, and HD groups was 58.7%, 53.8%, and 46.3%, respectively, while the corresponding PDR was 77.0%, 74.0%, and 61.2%. A significant increase in ADR, PDR, and SDR was observed between the Endocuff-AI and AI groups (ADR difference: 4.9%, 95% CI: 1.4%-8.2%, P = 0.03; PDR difference: 3.0%, 95% CI: 0.4%-5.8%, P = 0.04; SDR difference: 6.4%, 95% CI: 3.4%-9.7%, P < 0.01). Both Endocuff-AI and AI groups had a higher ADR, PDR, SDR, and advanced adenoma detection rate than the HD group (all P < 0.01). DISCUSSION: Endocuff in combination with AI further improves various colonic lesion detection rates when compared with AI alone.

Blue-light imaging or narrow-band imaging for proximal colonic lesions: a prospective randomized tandem colonoscopy study.

BACKGROUND AND AIMS: Blue-light imaging (BLI) is a new image-enhanced endoscopy with a wavelength filter similar to narrow-band imaging (NBI). We compared the 2 with white-light imaging (WLI) on proximal colonic lesion detection and miss rates. METHODS: In this 3-arm prospective randomized study with tandem examination of the proximal colon, we enrolled patients aged ≥40 years. Eligible patients were randomized in 1:1:1 ratio to receive BLI, NBI, or WLI during the first withdrawal from the proximal colon. The second withdrawal was performed using WLI in all patients. Primary outcomes were proximal polyp (pPDRs) and adenoma (pADRs) detection rates. Secondary outcomes were miss rates of proximal lesions found on tandem examination. RESULTS: Of 901 patients included (mean age, 64.7 years; 52.9% men), 48.1% underwent colonoscopy for screening or surveillance. The corresponding pPDRs of the BLI, NBI, and WLI groups were 45.8%, 41.6, and 36.6%, whereas the corresponding pADRs were 36.6%, 33.8%, and 28.3%. There was a significant difference in pPDR and pADR between BLI and WLI groups (difference, 9.2% [95% confidence interval {CI}, 3.3-16.9] and 8.3% [95% CI, 2.7-15.9]) and between NBI and WLI groups (difference, 5.0% [95% CI, 1.4-12.9] and 5.6% [95% CI, 2.1-13.3]). Proximal adenoma miss rates were significantly lower with BLI (19.4%) than with WLI (27.4%; difference, -8.0%; 95% CI, -15.8 to -.1) but not between NBI (27.2%) and WLI. CONCLUSIONS: Both BLI and NBI were superior to WLI on detecting proximal colonic lesions, but only BLI had lower proximal adenoma miss rates than WLI. (Clinical trial registration number: NCT03696992.).

Computer-assisted detection versus conventional colonoscopy for proximal colonic lesions: a multicenter, randomized, tandem-colonoscopy study.

BACKGROUND AND AIMS: Computer-assisted detection (CADe) is a promising technologic advance that enhances adenoma detection during colonoscopy. However, the role of CADe in reducing missed colonic lesions is uncertain. The aim of this study was to determine the miss rates of proximal colonic lesions by CADe and conventional colonoscopy. METHODS: This was a prospective, multicenter, randomized, tandem-colonoscopy study conducted in 3 Asian centers. Patients were randomized to receive CADe or conventional white-light colonoscopy during the first withdrawal of the proximal colon (cecum to splenic flexure), immediately followed by tandem examination of the proximal colon with white light in both groups. The primary outcome was adenoma/polyp miss rate, which was defined as any adenoma/polyp detected during the second examination. RESULTS: Of 223 patients (48.6% men; median age, 63 years) enrolled, 7 patients did not have tandem examination, leaving 108 patients in each group. There was no difference in the miss rate for proximal adenomas (CADe vs conventional: 20.0% vs 14.0%, P = .07) and polyps (26.7% vs 19.6%, P = .06). The CADe group, however, had significantly higher proximal polyp (58.0% vs 46.7%, P = .03) and adenoma (44.7% vs 34.6%, P = .04) detection rates than the conventional group. The mean number of proximal polyps and adenomas detected per patient during the first examination was also significantly higher in the CADe group (polyp: 1.20 vs .86, P = .03; adenoma, .91 vs .61, P = .03). Subgroup analysis showed that CADe enhanced proximal adenoma detection in patients with fair bowel preparation, shorter withdrawal time, and endoscopists with lower adenoma detection rate. CONCLUSIONS: This multicenter trial from Asia confirmed that CADe can further enhance proximal adenoma and polyp detection but may not be able to reduce the number of missed proximal colonic lesions. (Clinical trial registration number: NCT04294355.).

New insights on missed colonic lesions during colonoscopy through artificial intelligence-assisted real-time detection (with video).

BACKGROUND AND AIMS: Meta-analysis shows that up to 26% of adenomas could be missed during colonoscopy. We investigated whether the use of artificial intelligence (AI)-assisted real-time detection could provide new insights into mechanisms underlying missed lesions during colonoscopy. METHODS: A validated real-time deep-learning AI model for the detection of colonic polyps was first tested in videos of tandem colonoscopy of the proximal colon for missed lesions. The real-time AI model was then prospectively validated in a total colonoscopy in which the endoscopist was blinded to real-time AI findings. Segmental unblinding of the AI findings were provided, and the colonic segment was then re-examined when missed lesions were detected by AI but not the endoscopist. All polyps were removed for histologic examination as the criterion standard. RESULTS: Sixty-five videos of tandem examination of the proximal colon were reviewed by AI. AI detected 79.1% (19/24) of missed proximal adenomas in the video of the first-pass examination. In 52 prospective colonoscopies, real-time AI detection detected at least 1 missed adenoma in 14 patients (26.9%) and increased the total number of adenomas detected by 23.6%. Multivariable analysis showed that a missed adenoma(s) was more likely when there were multiple polyps (adjusted odds ratio, 1.05; 95% confidence interval, 1.02-1.09; P < .0001) or colonoscopy was performed by less-experienced endoscopists (adjusted odds ratio, 1.30; 95% confidence interval, 1.05-1.62; P = .02). CONCLUSIONS: Our findings provide new insights on the prominent role of human factors, including inexperience and distraction, on missed colonic lesions. With the use of real-time AI assistance, up to 80% of missed adenomas could be prevented. (Clinical trial registration number: NCT04227795.).

Accuracy of artificial intelligence-assisted detection of upper GI lesions: a systematic review and meta-analysis.

BACKGROUND AND AIMS: Artificial intelligence (AI)-assisted detection is increasingly used in upper endoscopy. We performed a meta-analysis to determine the diagnostic accuracy of AI on detection of gastric and esophageal neoplastic lesions and Helicobacter pylori (HP) status. METHODS: We searched Embase, PubMed, Medline, Web of Science, and Cochrane databases for studies on AI detection of gastric or esophageal neoplastic lesions and HP status. After assessing study quality using the Quality Assessment of Diagnostic Accuracy Studies tool, a bivariate meta-analysis following a random-effects model was used to summarize the data and plot hierarchical summary receiver-operating characteristic curves. The diagnostic accuracy was determined by the area under the hierarchical summary receiver-operating characteristic curve (AUC). RESULTS: Twenty-three studies including 969,318 images were included. The AUC of AI detection of neoplastic lesions in the stomach, Barrett's esophagus, and squamous esophagus and HP status were .96 (95% confidence interval [CI], .94-.99), .96 (95% CI, .93-.99), .88 (95% CI, .82-.96), and .92 (95% CI, .88-.97), respectively. AI using narrow-band imaging was superior to white-light imaging on detection of neoplastic lesions in squamous esophagus (.92 vs .83, P < .001). The performance of AI was superior to endoscopists in the detection of neoplastic lesions in the stomach (AUC, .98 vs .87; P < .001), Barrett's esophagus (AUC, .96 vs .82; P < .001), and HP status (AUC, .90 vs .82; P < .001). CONCLUSIONS: AI is accurate in the detection of upper GI neoplastic lesions and HP infection status. However, most studies were based on retrospective reviews of selected images, which requires further validation in prospective trials.

Targeted genomic profiling identifies frequent deleterious mutations in FAT4 and TP53 genes in HBV-associated hepatocellular carcinoma.

BACKGROUND: Hepatitis B virus (HBV) is the major risk factor for hepatocellular carcinoma (HCC). The molecular mechanisms underlying HBV-associated HCC pathogenesis is still unclear. Genetic alterations in cancer-related genes have been linked to many human cancers. Here, we aimed to explore genetic alterations in selected cancer-related genes in patients with HBV-associated HCC. METHODS: Targeted sequencing was used to analyze six cancer-related genes (PIK3CA, TP53, FAT4, IRF2, HNF4α and ARID1A) in eight pairs of HBV-associated HCC tumors and their adjacent non-tumor tissues. Sanger sequencing, quantitative PCR, Western-blotting and RNAi-mediated gene knockdown were used to further validate findings. RESULTS: Targeted sequencing revealed thirteen non-synonymous mutations, of which 9 (69%) were found in FAT4 and 4 (31%) were found in TP53 genes. Non-synonymous mutations were not found in PIK3CA, IRF2, HNF4α and ARID1A. Among these 13 non-synonymous mutations, 12 (8 in FAT4 and 4 in TP53) were predicted to have deleterious effect on protein function by in silico analysis. For TP53, Y220S, R249S and P250R non-synonymous mutations were solely identified in tumor tissues. Further expression profiling of FAT4 and TP53 on twenty-eight pairs of HCC tumor and non-tumor tissues confirmed significant downregulation of both genes in HCC tumors compared with their non-tumor counterparts (P < 0.001 and P < 0.01, respectively). Functional analysis using RNAi-mediated knockdown of FAT4 revealed an increased cancer cell growth and proliferation, suggesting the putative tumor suppressor role of FAT4 in HCC. CONCLUSIONS: This study highlights the importance of FAT4 and TP53 in HCC pathogenesis and identifies new genetic variants that may have potentials for development of precise therapy for HCC.

Pilot model of hepatitis C virus micro-elimination in high-risk populations in Hong Kong: Barriers and facilitators.

BACKGROUND: Although the general seroprevalence of hepatitis C virus (HCV) infection in Hong Kong is <0.5 %, Hong Kong is still striving for HCV elimination owing to barriers in care cascade encompassing linkage-to-care (LTC), treatment initiation and adherence. We aimed to evaluate the feasibility of a pilot model of micro-elimination to strengthen the HCV care cascade for high-risk groups in Hong Kong. METHODS: We initiated the pilot Conquering Hepatitis vIa Micro-Elimination (CHIME) program which adopts an integrated care approach involving outreach visits to halfway house or drug rehabilitation centers run by non-governmental organizations. Participants with history of injection drug use (PWID), recreational drug use, or imprisonment were included. We performed point-of-care test for anti-HCV with reflex HCV RNA testing. LTC with government-subsidized direct acting antiviral was provided to viremic participants. We compared the impact on the care cascade with a cohort of HCV patients (17.8 % PWID) under usual care. RESULTS: 396 participants (62.9 % PWID) were screened and 187 (47.2 %) were viremic, of which 29.8 % had cirrhosis. Proportion with LTC, treatment initiation and adherence were 76.5 % and 63.7 %, 90.9 % and 85.8 %, and 90.0 % and 92.2 %, for the CHIME program and usual care, respectively. The CHIME program was significantly associated with higher odds of LTC (OR 1.797, 95 % CI 1.221-2.644). Non-engagement in care (affecting 37.9 % participants with HCV viremia) was associated with unemployment (OR 2.165, 95 % CI 1.118-4.190). CONCLUSION: The pilot CHIME program demonstrated feasibility of an integrated approach to consolidate the HCV care cascade in high-risk populations in Hong Kong.

Longitudinal renal changes in chronic hepatitis B patients treated with entecavir versus TDF: a REAL-B study.

BACKGROUND AND AIMS: We aimed to compare the longitudinal changes in estimated glomerular filtration rate (eGFR) in chronic hepatitis B (CHB) patients treated with entecavir (ETV) vs. tenofovir disoproxil fumarate (TDF). METHODS: This is a retrospective study of 6189 adult treatment-naïve CHB patients initiated therapy with TDF (n = 2482) or ETV (n = 3707) at 25 international centers using multivariable generalized linear modeling (GLM) to determine mean eGFR (mL/min/1.73 m2) and Kaplan-Meier method to estimate incidence of renal impairment (≥ 1 chronic kidney disease [CKD] stage worsening). We also examined above renal changes in matched ETV and TDF patients (via propensity score matching [PSM] on age, sex, diabetes mellitus [DM], hypertension [HTN], cirrhosis, baseline eGFR, and follow-up duration). RESULTS: In the overall cohort (mean age 49.7 years, 66.2% male), the baseline eGFR was higher for TDF vs. ETV group (75.9 vs. 74.0, p = 0.009). PSM yielded 1871 pairs of ETV or TDF patients with baseline eGFR ≥ 60 and 520 pairs for the eGFR < 60 group. GLM analysis of the overall (unmatched) cohort and PSM cohorts revealed lower adjusted mean eGFRs in TDF (vs. ETV) patients (all p < 0.01) during 10 years of follow-up. Among PSM eGFR ≥ 60 patients, the 5-year cumulative incidences of renal impairment were 42.64% for ETV and 48.03% for TDF (p = 0.0023). In multivariable Cox regression, TDF vs. ETV (adjusted HR 1.26, 95% CI 1.11-1.43) was associated with higher risk of worsening renal function. CONCLUSION: Over the 10-year study follow-up, compared to ETV, TDF was associated with a lower mean eGFR and higher incidence of renal impairment.

Impact of first wave of COVID-19 on outcomes of hospitalization for upper gastrointestinal bleeding in Hong Kong: a population-based study.

Background and study aims The COVID-19 pandemic has caused a major disruption in the healthcare system. This study determined the impact of the first wave of COVID-19 on the number and outcome of patients hospitalized for upper gastrointestinal bleeding (UGIB) in Hong Kong. Patients and methods Records of all patients hospitalized for UGIB in Hong Kong public hospitals between October 2018 and June 2020 were retrieved. The number and characteristics of patients hospitalized for UGIB after COVID-19 was compared by autoregressive integrated moving average (ARIMA) model prediction and historical cohort. Results Since the first local case of COVID-19, there was an initial drop in UGIB hospitalizations (observed 29.8 vs predicted 35.5 per week; P  = 0.05) followed by a rebound (39.8 vs 26.7 per week; P  < 0.01) with a turning point at week 14 (Petitt's test, P  < 0.001). There was a negative association between the number of COVID-19 cases and the number of patients hospitalized for UGIB (Pearson correlation -0.53, P  < 0.001). Patients admitted after the outbreak of COVID-19 had lower hemoglobin (7.5 vs baseline 8.3 g/dL; P  < 0.01) and a greater need for blood transfusion (64.5 % vs baseline 50.4 %; P  < 0.01), but similar rates of all-cause mortality (6.9 % vs 7.1 %; P  = 0.82) and rebleeding (6.7 % vs 5.1 %; P  = 0.11). There was also a higher proportion of patients with variceal bleeding (10.5 % vs baseline 5.3 %; P  < 0 .01). Conclusions There was a dynamic change in the number of patients hospitalized for UGIB in Hong Kong during the first wave of the COVID-19 outbreak, with more obvious impact during the initial phase only.

Feedback from artificial intelligence improved the learning of junior endoscopists on histology prediction of gastric lesions.

Background and study aims Artificial intelligence (AI)-assisted image classification has been shown to have high accuracy on endoscopic diagnosis. We evaluated the potential effects of use of an AI-assisted image classifier on training of junior endoscopists for histological prediction of gastric lesions. Methods An AI image classifier was built on a convolutional neural network with five convolutional layers and three fully connected layers A Resnet backbone was trained by 2,000 non-magnified endoscopic gastric images. The independent validation set consisted of another 1,000 endoscopic images from 100 gastric lesions. The first part of the validation set was reviewed by six junior endoscopists and the prediction of AI was then disclosed to three of them (Group A) while the remaining three (Group B) were not provided this information. All endoscopists reviewed the second part of the validation set independently. Results The overall accuracy of AI was 91.0 % (95 % CI: 89.2-92.7 %) with 97.1 % sensitivity (95 % CI: 95.6-98.7%), 85.9 % specificity (95 % CI: 83.0-88.4 %) and 0.91 area under the ROC (AUROC) (95 % CI: 0.89-0.93). AI was superior to all junior endoscopists in accuracy and AUROC in both validation sets. The performance of Group A endoscopists but not Group B endoscopists improved on the second validation set (accuracy 69.3 % to 74.7 %; P  = 0.003). Conclusion The trained AI image classifier can accurately predict presence of neoplastic component of gastric lesions. Feedback from the AI image classifier can also hasten the learning curve of junior endoscopists in predicting histology of gastric lesions.